Abstract
Spectral unmixing is an important problem for remotely sensed hyperspectral data exploitation. Automatic spectral unmixing can be viewed as a three-stage problem, where the first stage is subspace identification, the next one is endmember extraction, and the final one is abundance estimation. In this sequence, endmember extraction is the most challenging problem. Many researchers have attempted to extract endmembers from hyperspectral images using spectral information only. However, it is well known that the inclusion of spatial information can improve the endmember extraction task. In this article, we introduce a new endmember extraction algorithm that exploits both spectral and spatial information. A main innovation of the proposed algorithm is that spatial information is exploited using entropy, while spectral information is exploited using convex set optimization. In the literature, none of the spatial–spectral algorithms has used entropy as spatial information. The inclusion of this entropy-based spatial information improves the accuracy of the endmember extraction process. The results obtained by the proposed algorithm are compared (using a variety of metrics) with those obtained by other state-of-the-art methods, using both synthetic and real datasets. Our experimental results demonstrate that the proposed algorithm outperforms many available algorithms.
Highlights
R EMOTE sensing is used in various applications of Earth science, geography, land surveying, and Mars exploration [1]
The results obtained by the proposed algorithm are compared to those obtained by ten well-known endmember extraction algorithms (SVMAX, alternating volume maximization (AVMAX), vertex component analysis (VCA), TRIple-P:P-norm based pure pixel identification (TRIP), pixel purity index (PPI), Independent component analysis (ICA), automatic morphological endmember extraction (AMEE), spatial preprocessing for endmember extraction (SPEE), region-based spatial preprocessing (RBSPP), and spatial–spectral preprocessing (SSPP)) on both synthetic and real datasets
An innovative characteristic of the proposed method is that it combines the concept of band entropy and convex geometry
Summary
R EMOTE sensing is used in various applications of Earth science, geography, land surveying, and Mars exploration [1]. Most works assume the linear mixing model, as it is a simple approximation to real-world applications [2] In this model, the concept of endmember is a key aspect, since endmembers are spectrally distinct signatures of pure materials that can be used to model (linearly or nonlinearly) the mixed pixels in the scene. The SPEE algorithm spatially weighs the spectral information related to each pixel for endmember extraction. We develop a new algorithm for endmember extraction that combines both spectral and spatial information. ICA uses the entropy of various spectra as a spectral feature, while the proposed algorithm uses the entropy of each band as a spatial feature At this point, it is important to emphasize that many hyperspectral endmember extraction algorithms [16]–[18], [24]–[26] have used the concept of convex geometry optimization.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
More From: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.